Very cool! Function calling seems to be a new paradigm that is really taking off.

Does anyone know how the LLM vendors are actually implementing function calling? Is it just a thoughtful prompt and a loop where they parse outputs and check if it corresponds to the arguments of the function? Or something else?

Very useful when parsing long reams of data with prompts built in. I can think of using a Incident responder using this intelligently for searching security event logs for a malicious user-id